没有合适的资源?快使用搜索试试~ 我知道了~
ITK_Registration.pdf
1星 需积分: 44 11 下载量 49 浏览量
2020-06-02
18:03:13
上传
评论
收藏 3.51MB PDF 举报
温馨提示
试读
27页
ITK-SNAP配准原理介绍,主要讲解自动配准的基本原理,实现方法和基本算法,方便研究者掌握医学图像配准的原理方法,以及二次开发!
资源推荐
资源详情
资源评论
2/9/20
1
This work by John Galeotti and Damion Shelton, © 2004-2020, was made possible in part by NIH NLM contract#
HHSN276201000580P, and is licensed under a Creative Commons Attribution 3.0 Unported License. To view a copy of this
license, visit http://creativecommons.org/licenses/by/3.0/ or send a letter to Creative Commons, 171 2nd Street, Suite 300, San
Francisco, California, 94105, USA. Permissions beyond the scope of this license may be available by emailing itk@galeotti.net.
The most recent version of these slides may be accessed online via http://itk.galeotti.net/
Methods in Medical Image Analysis - Spring 2020
16-725 (CMU RI) : BioE 2630 (Pitt)
Dr. John Galeotti
Based in part on Damion Shelton’s slides from 2006
Lecture 8
Registration with ITK
1
1
For more info/gory detail…
§ Please see the following for exhaustive detail:
§ Chapter 3 in the ITK Software Guide Book 2
§ Insight into Images
§ ITK Source Tree
§ Examples/RegistrationITKv4/
§ E.g. Examples/RegistrationITKv4/ImageRegistration1.cxx
§ ITK Doxygen
§ http://www.itk.org/Doxygen50/html/group__RegistrationFilters.html
§ http://www.itk.org/Doxygen50/html/group__Group-Registration.html
§ http://www.itk.org/Doxygen50/html/group__Group-Numerics.html
§ SimpleITK:
§ http://insightsoftwareconsortium.github.io/SimpleITK-Notebooks/
§ See all the Python Registration (6x) notebooks, especially:
§ http://insightsoftwareconsortium.github.io/SimpleITK-
Notebooks/Python_html/60_Registration_Introduction.html
§ https://itk.org/SimpleITKDoxygen/html/classitk_1_1simple_1_1ImageRegistrati
onMethod.html
2
2
2/9/20
2
What is registration?
§The process of aligning a target image to a
source image
§More generally, determining the transform that
maps points in the target image to points in the
source image
3
3
Transform types
§Rigid (rotate, translate)
§Affine (rigid + scale & shear)
§Deformable = non-rigid (affine + vector field)
§Many others
4
4
2/9/20
3
Registration in ITK
§ ITK uses an extensible registration framework
§ Various interchangeable classes exist
§ Relatively easy to “twiddle” the part you’re interested in while recycling
prior work
§ The newer ITKv4 Registration framework is separate from the legacy
framework.
§ The legacy framework follows traditional practice
§ Version 4 registration is more flexible and thus more complex
§ Use the v4 framework whenever practical
§ SimpleITK recently gained registration support at v. 0.9, improved in v1.0
§ For “simplified” complex registration, consider using ANTS instead:
§ http://www.picsl.upenn.edu/ANTS/
§ http://stnava.github.io/ANTs/
5
5
New in ITKv4
(ImageRegistrationMethodv4, etc.)
§ New unified, improved, and fully multi-threaded optimization and
registration framework (including multi-threaded metrics)
§ Dense deformation fields (including a new transform that encapsulates a
dense deformation field)
§ Point Set registration methods (landmark or label guided registration)
§ Automatic parameter scale estimation for transforms
§ Automatic step-size selection for gradient-based registration optimizers
§ Composite Transforms (grouping multiple transforms into a single one)
§ Symmetric registration (where the Fixed and Moving images make
unbiased contributions to the registration)
§ New metrics for Demons and Mutual Information
§ Diffeomorphic (velocity field) deformable registration
§ Additional evolutionary optimizers
§ Improved B-Spline registration approach available and bug fixes to old
framework
§ Accurately transform and reorient covariant tensors and vectors
6
List taken from http://www.itk.org/Wiki/ITK_Release_4/Why_Switch_to_ITKv4 and
http://www.itk.org/Wiki/ITK_Release_4/Migration_Plan/Release_Notes/ITKv4_Final_Release_Notes
6
2/9/20
4
ITKv4 Registration
§ Uses a different framework than “traditional” ITK
registration. The new framework is designated with a
“v4” suffix.
§ You must use a v4 metric and a v4 optimizer when
doing a v4 registration!
§ Take a look here:
http://www.itk.org/Doxygen50/html/group__ITKRegistrationMethodsv4.html
http://www.itk.org/Doxygen50/html/group__ITKMetricsv4.html
http://www.itk.org/Doxygen50/html/group__ITKOptimizersv4.html
ITK source code: Modules/Registration/RegistrationMethodsv4/include/
ITK source code: Modules/Registration/Metricsv4/include/
ITK source code: Modules/Numerics/Optimizersv4/include/
§ Pay special attention to:
§ MattesMutualInformationImageToImageMetricv4
§ DemonsImageToImageMetricv4
§ QuasiNewtonOptimizerv4 (an improved gradient descent)
7
7
Typical registration terminology
§Fixed image f(x) - stationary in space
§Moving image m(x) - the fixed image with an
unknown transform applied
§Goal: recover the transform T(x) which maps
points in f(x) to m(x)
8
8
2/9/20
5
Typical registration framework
§2 input images, fixed and moving
§Metric - determines the “fitness” of the current
registration iteration
§Optimizer - adjusts the transform in an attempt
to improve the metric
§Interpolator - applies transform to image and
computes sub-pixel values
9
9
Figure 8.2 from the ITK Software Guide v 2.4, by Luis Ibáñez, et al.
10
Typical registration flowchart
10
剩余26页未读,继续阅读
资源评论
- huaxia0482020-10-22说的特别笼统,没什么有用的信息
临海观风
- 粉丝: 1
- 资源: 28
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功